Need help with training priviledged teach policy

[songlin]

Hi Tairan,

Thanks for sharing the code.

I tried to reproduce the results of the privileged teacher policy with the following scripts:

python legged_gym/scripts/train_hydra.py \
  --config-name=config_teleop \
  task=h1:teleop run_name=OmniH2O_TEACHER \
  env.num_observations=913 \
  env.num_privileged_obs=990 \
  motion.teleop_obs_version=v-teleop-extend-max-full \
  motion=motion_full \
  motion.extend_head=True \
  num_envs=4096 \
  asset.zero_out_far=False \
  asset.termination_scales.max_ref_motion_distance=1.5 \
  sim_device=cuda:0 \
  motion.motion_file=resources/motions/h1/amass_phc_filtered.pkl \
  rewards=rewards_teleop_omnih2o_teacher \
  rewards.penalty_curriculum=True \
  rewards.penalty_scale=0.5

After training about 200K steps, I still observed very bad results when playing the teach policy, nearly 0 success rate:

Loaded 1 motions with a total length of 9.300s and 280 frames.                                              | 0/1 [00:00<?, ?it/s]
Terminated: 0 | max frames: 467 | steps 12 | Start: 62 | Succ rate: 0.016 | Mpjpe: 252.459

The video results:

Screencast.from.09-13-2025.10.44.09.AM.webm

Here is my training curve:

What might I be doing wrong?

Thank you.

[liyongjian2023]

Which motion does the video show? I encountered similar problem when the robot are kicking. The robot cannot perform well.